The object of this project is to compute the conformations of segments of double and single stranded DNA modified at a single central site by various covalently linked polycyclic aromatic carcinogens. The calculated conformations will be compared to available data on mutagenicity and carcinogenicity of the particular lesion, in order to seek a general relationship between deformations in DNA and the biological responses to them. The base sequences to be investigated will be largely those that have been implicated as mutagenic sequences in bacterial systems, including especially the alternating G-C's. Our computational method proceeds in distinct phases. First we calculate the most favorable conformations of a carcinogen modified dinucleoside-monophosphate, using minimized semi- empirical potential energy methods with about 4000 trails. The low energy conformations from this stage are then incorporated in single stranded tetramers with the added residues in the A, B, Z or other helical conformation, and the energy is again minimized. A selection is again made on energetic grounds, and the lowest energy forms are further incorporated into larger polymers for final energy minimization. New search strategies will be devised and tested in addition to the above, and employed if proven useful. We will search for novel unmodified single stranded polymers by combining stepwise the low energy dimer subunits and minimizing the energy. Novel unmodified duplexes will be sought from the single stranded forms. The important computed structures will then be employed in addition to the A, B and Z helices in the carcinogen studies. Furthermore, we will investigate single stranded trimers, modified at the central based, as a conformational building block. Unusual carcinogen modified duplex or partly denatured forms, suggested from the dimer and trimer studies, such as bulges, cruciforms and B-Z junctions, will be sought both computationally and with the Evans and Sutherland Interactive Computer Graphics System. In the above studies, we shall employ potentials and programs that already have been developed and are in working order. At the same time, we will also work on some basic areas of methodology, including the development of a new cartesian space program, specific incorporation of metals ions and the investigations of problems associated with the minimization method. Furthermore, we may be able to proceed to molecular dynamics in the proposed project period, which may enable us to locate important new forms and paths between conformations that are significant for carcinogen-modified DNA.